Snow HYDRO-CH2018 SYNTHESIS REPORT CHAPTERS: "FUTURE CHANGES IN HYDROLOGY" - DORA 4RI

Snow
HYDRO-CH2018 SYNTHESIS REPORT CHAPTERS: “FUTURE
CHANGES IN HYDROLOGY“

C. MARTY, M. BAVAY, D.FARINOTTI, AND M. HUSS

IM AUFTRAG DES BUNDESAMTES FÜR UMWELT BAFU – FEBRUAR 2020

EINE STUDIE IM RAHMEN DES NCCS THEMENSCHWERPUNKTES “HYDROLOGISCHE
GRUNDLAGEN ZUM KLIMAWANDEL” DES NATIONAL CENTRE FOR CLIMATE SERVICES

Impressum

Commissioned by: Federal Office for the Environment (FOEN), Hydrology Division, CH-3003
Bern. The FOEN is an agency of the Federal Department of the Environment, Transport,
Energy and Communications (DETEC)

Contractor: ETH Zürich, Versuchsanstalt für Wasserbau, Hydrologie und Glaziologie (VAW),
HIA C58, Hönggerbergring 26, CH-8093 Zürich

Authors: Christoph Marty(3), Matthias Bavay(3), Daniel Farinotti (1,2) and Matthias Huss(1,2,4)
(1)
      Laboratory of Hydraulics, Hydrology and Glaciology (VAW), ETH Zurich
(2)
      Swiss Federal Institute for Forest, Snow and Landscape Research (WSL)
(3)
      WSL Institute for Snow and Avalanche Research (SLF)
(4)
      Department of Geosciences, University of Fribourg (UniFR)

FOEN support: Fabia Huesler, Petra Schmocker-Fackel

Note: This study/report was prepared under contract to the Federal Office for the Environment
(FOEN). The contractor bears sole responsibility for the content.

Citation: Marty, Ch., Bavay, M., Farinotti, A., and Huss, M. 2020. Snow: Hydro-CH2018
synthesis report chapters: “future changes in hydrology“. Hydro-CH2018 Project.
Commissioned by the Federal Office for the Environment (FOEN), Bern, Switzerland. 25 pp,
doi: 10.16904/slf.2
DOI: 10.16904/slf.2

Summary
An important feature of snow cover is the fact that its volume and duration is subject to large
year-to-year fluctuations. As frozen precipitation, snow cover is a natural water reservoir that
delays precipitation to runoff and is thus of outstanding importance for the seasonal water
balance in Switzerland. Over a whole year, approximately 40% (22 km3) of the annual runoff
currently comes from snow melting and only 1-2 % from glacier melting. Typically, the snow
cover in the Alpine region builds up over the autumn and winter months, reaches its
maximum between February and May, depending on the altitude, and dominates the runoff
processes during melting in the following spring and summer months.
Due to the great dependence on minus temperatures and precipitation, the snow cover
reacts sensitively to temperatures above 0° Celsius and more or less precipitation. Due to
climate change and the associated warming, the proportion of precipitation that falls as snow
decreases measurably. In addition to this reduction in snowfall, the warmer temperatures
also cause the snow cover to melt more quickly. The decline in snowfall has so far mainly
affected lower altitudes, where winter temperatures often reach positive levels. The
proportion of days with snowfall below 500 m above sea level has fallen by approx. 40%
since 1961 and the water temporarily stored in the snow (snow water equivalent) in spring
below 1000 m asl. even by approx. 75%.
As climate change progresses, this trend is likely to continue and above all affect higher
zones. Even at higher altitudes, the snow cover will then start later, melt away earlier and the
area of permanent snow cover will strongly diminish. Due to the projected increase in winter
precipitation by the end of the century, the snow cover above 2500 m above sea level will
decrease only slightly until the start of the snow melt period. All in all, however, the water
stored in the snow will clearly decrease. This development will also have an effect on the
water bodies. Today nival regimes, i.e. regimes shaped by snow, are shifting towards pluvial
regimes, i.e. regimes dominated by rain. Overall, winter runoff increases, summer runoff
decreases. By the end of the century, the proportion of runoff from snowmelt will decrease
throughout Switzerland, albeit to a lesser extent than the proportion from glacier melt.

                                                I

Zusammenfassung

Ein wichtiges Merkmal der Schneebedeckung ist die Tatsache, dass deren Volumen und Dauer
grossen Jahr-zu-Jahr Schwankungen unterliegt. Als gefrorener Niederschlag ist die
Schneedecke ein natürlicher Wasserspeicher, der den Niederschlag verzögert zum Abfluss bringt
und damit eine herausragende Bedeutung für den saisonalen Wasserhaushalt in der Schweiz
hat. Über ein ganzes Jahr betrachtet, stammen aktuell ungefähr 40% (ca. 22 km3) des jährlichen
Abflusses aus der Schneeschmelze und nur gerade 1-2 % aus der Gletscherschmelze.
Typicherweise baut sich die Schneedecke im Alpenraum über die Herbst- und Wintermonate
hinweg auf, erreicht ihr Maximum je nach Höhenlage zw. Februar und Mai und dominiert während
des Abschmelzens in den folgenden Frühlings- und Sommermonaten die Abflussprozesse.

Aufgrund der grossen Abhängigkeit von Minus-Temperaturen und Niederschlag reagiert die
Schneedecke empfindlich auf Temperaturen oberhalb 0° Celsius und mehr oder weniger
Niederschlag. Auf Grund des Klimawandels und der damit verbundenen Erwärmung nimmt der
Anteil des Niederschlages, welcher als Schnee fällt, messbar ab. Neben diesem Rückgang der
Schneefälle schmilzt die Schneedecke auf Grund der wärmeren Temperaturen zudem schneller
weg. Betroffen vom Schneerückgang waren bisher vor allem tiefere Lagen, weil dort die
Wintertemperaturen häufig positive Temperaturen erreichen. So ist der Anteil der Tage mit
Schneefall unter 500 m ü.M. seit 1961 um ca. 40 % zurückgegangen und das im Frühling im
Schnee zwischengespeicherte Wasser (Schneewasseräquivalent) unterhalb von 1000 m ü.M.
gar um ca. 75%.

Mit fortschreitendem Klimawandel dürfte sich dieser Trend weiter fortsetzen und vor allem auch
höhere Zonen betreffen. Auch in den hohen Lagen setzt dann die Schneedecke später ein,
schmilzt früher weg und die Fläche mit permanenter Schneedecke wird stark zurückgehen.
Aufgrund der bis Ende Jahrhundert projizierten Zunahme der Winterniederschläge nimmt die
Schneedecke oberhalb 2500 m bis zum Einsetzen der Schneeschmelze nur leicht ab. Insgesamt
wird das im Schnee gespeicherte Wasser aber klar zurückgehen. Diese Entwicklung wirkt sich
natürlich auf die Gewässer aus. Heute nival, also vom Schnee geprägte Regime verschieben sich
hin zu pluvialen, d.h. vom Regen dominierten Regimetypen. Die Winterabflüsse nehmen
insgesamt zu, die Sommerabflüsse ab. Bis Ende Jahrhundert nimmt der Anteil Abfluss aus der
Schneeschmelze schweizweit ab, wenn auch klar weniger stark als der Anteil aus der
Gletscherschmelze.

                                              II

Résumé

Une caractéristique importante de la couverture neigeuse est que son volume et sa durée sont
soumis à d'importantes fluctuations d'une année à l'autre. En accumulant les précipitations
solides, la couverture neigeuse se comporte comme un réservoir d'eau naturel qui retarde le
ruissellement des précipitations et est donc d'une importance capitale pour le bilan hydrologique
saisonnier en Suisse. En moyenne, environ 40 % (22 km3) du ruissellement annuel provient
actuellement de la fonte des neiges et seulement 1-2 % de la fonte des glaciers. Dans l'espace
alpin la couverture neigeuse s'accumule généralement au cours des mois d'automne et d'hiver,
atteint son maximum entre février et mai selon l'altitude et domine les processus de ruissellement
pendant la fonte au printemps et à l'été suivants.

Étant très dépendante aux températures négatives et aux précipitations, la couverture neigeuse
réagit de manière sensible aux températures supérieures à 0° Celsius et à la quantité de
précipitations. En raison des changements climatiques et du réchauffement connexe, la
proportion des précipitations qui tombent sous forme de neige diminue de façon mesurable. En
plus de cette réduction des chutes de neige, les températures plus chaudes font fondre plus
rapidement la couverture neigeuse. Jusqu'à présent, sa diminution a surtout touché les basses
altitudes où les températures hivernales atteignent souvent des valeurs positives. Depuis 1961,
le nombre de jours avec des chutes de neige en dessous de 600 m d'altitude a diminué d'environ
40 % et l'eau stockée dans la couverture neigeuse saisonnière (équivalent en eau de la neige)
au printemps en dessous de 1000 m d'altitude, d'environ 75 %.

Au fur et à mesure que le changement climatique s’intensifie, cette tendance est susceptible de
se poursuivre et surtout d'affecter des zones plus élevées. Même à ces altitudes, la couverture
neigeuse commencera plus tard, fondra plus tôt et la zone de couverture de neige permanente
diminuera fortement. En raison de l'augmentation prévue des précipitations hivernales jusqu’à la
fin du siècle, la couverture neigeuse au-dessus de 2500 m d'altitude ne diminuera que légèrement
jusqu'à la période de fonte. Dans l'ensemble, cependant, l’équivalent en eau de la neige va
clairement diminuer. Ce développement aura naturellement un effet sur les plans d'eau et rivières.
Aujourd'hui, les régimes hydrologiques nivaux, c'est-à-dire les régimes hydrologiques dominés
par la neige, se transforment en régimes pluviaux, c'est-à-dire des régimes dominés par la pluie.
Le ruissellement hivernal augmente globalement, le ruissellement estival diminue. Jusqu’à la fin
du siècle, la contribution hydrologique de l’équivalent en eau de la neige diminuera en Suisse,
mais dans une moindre mesure que celle des eaux de fonte glaciaire.

                                               III

Table of contents
List of abbreviations .............................................................................................................. V
1.      Introduction .................................................................................................................... 1
     1.1      Snow cover and climate change .............................................................................. 1
     1.2 The role of snow for hydrology in Switzerland .............................................................. 2
     1.3 Objectives of this chapter ............................................................................................. 3
2.      Review of recent reports................................................................................................. 4
     2.1 Snow declines and its consequences as seen in recent reports ................................... 4
     2.2 Open questions in the reviewed reports ....................................................................... 6
     3.1 State of literature.......................................................................................................... 8
     3.2 Recent advances ......................................................................................................... 9
     3.3 New technologies and remote sensing products .........................................................10
4.      Implications of projected snow changes on Swiss water resources ...............................13
     4.1 Water availability .........................................................................................................13
     4.2 Hydropower production ...............................................................................................13
     4.3 Natural hazards and other impacts..............................................................................14
5.      Research gaps and open questions ..............................................................................16
6.      Translating research findings into policy-relevant messages .........................................18

                                                                    IV

List of abbreviations
ACQWA (project): Assessing Climate Impacts on the Quantity and quality of Water.
asl: above sea level.
CCHydro (project): Climate Change and Hydrology in Switzerland.
CLISP (project): Climate Change Adaptation by Spatial Planning in the Alpine Space.
CHR (institution): International Commission for the Hydrology of the Rhine basin.
CHy (institution): Swiss Hydrological Commission.
CH2011 (project): Climate change scenarios downscaled to the Swiss domain in 2011.
CH2014-impacts (project): Impacts of climate change in Switzerland on different sectors.
CH2018 (project): Climate change scenarios downscaled to the Swiss domain in 2018.
CORDEX (project): Coordinated Regional Climate Downscaling Experiment.
CO2: Carbon Dioxide.
DEM (technical term): Digital Elevation Model.
ESA (institution): European Space Agency.
FOEN (institution): Federal Office for the Environment.
FUGE (project): Future glacier evolution and consequences for the hydrology.
GCM (technical term): Global Climate Model.
GERM (model): Glacier Evolution Runoff Model.
GLAMOS: Glacier Monitoring in Switzerland.
GPS (technical term): Global Positioning System.
IGOS (project): Integrated Global Observing Strategy.
LIDAR (method): Light Detection And Ranging.
MODIS (satellite): Moderate Resolution Imaging Spectroradiometer.
NAO (technical term): North Atlantic Oscillation.
NCCR Climate (institution): National Centre of Competence in Research "Climate Variability,
Predictability and Climate Risks".
NCCS (institution): National Centre for Climate Services.
OCCR (institution): Oeschger Centre for Climate Change Research.
PREVAH (model): PREecipitation-Runoff-EVApotranspiration HRU Model.
RCP (technical term): Representative Concentration Pathways.
SCCER-SoE (institution): Swiss Competence Centre for Energy Research, Supply of
Electricity.
SGHL (institution): Schweizerische Gesellschaft für Hydrologie und Limnologie.
SMP (instrument): SnowMicroPen.
SNSF (institution): Swiss National Science Foundation.
SRES (technical term): Special Report on Emissions Scenarios (e.g. A1B or A2 scenario).
SWE: Snow Water Equivalent.
UAV (technical term): Unmanned Aerial Vehicle

                                            V

1. Introduction
1.1 Snow cover and climate change
Terrestrial snow has the largest geographic extent of all cryospheric components. It covers
nearly 50 million km2 of the northern hemisphere in winter. Snow provides a natural reservoir
of fresh water by storing precipitation and delaying runoff. Snowmelt fills rivers and aquifers
with water for drinking, food and energy production, and transportation. On the other hand,
snowmelt floods and avalanches claim lives and property every year. The influence of snow
extends from engineering, accessibility for humans and wildlife to recreation and many other
aspects of daily living with significant economic implications (Hock et al. 2019, IGOS 2007, ).
The timing, volume or extent of snow on the ground is always the result of an interplay between
temperature and precipitation. E.g. snowfall can only be observed when precipitation is
occurring during cold enough temperatures. A hereby built up snow cover can only last as long
as temperatures are suitable to prevent melt. This implies that a deep snow pack depends on
high precipitation amounts and sufficiently low temperatures. A snowpack therefore always
represents the cumulative effects of snow accumulation and ablation during the preceding
months. Seasonal snow acts as a control on summer soil-water storage and variations in
snowpack are a key component in the global heat budget. Fresh snow has a high albedo and
reductions in snow cover due to warming and increasing amount of black carbon in the snow
are expected to contribute to polar amplification of climate change (Phitan et al. 2014).
This high sensitivity of snow to changes in temperature and precipitation makes it a primary
indicator of climate change and the corresponding impacts on the redistribution and
acceleration of the water cycle. Several studies demonstrate that the climate change induced
temperature increase already reduces snow accumulation as a greater fraction of the
precipitation falls as rain (Serquet et al. 2011), while warmer spring and summer temperatures
intensify the spring snow melt in the Alps and thus cause a shift to earlier snow disappearance
dates (Marty and Meister 2012).
In the Alps the snow depth and snow duration show negative trends over the past decades.
The changes are typically elevation dependent, with more (less) pronounced changes at low
(high) elevations (Marty 2008; Durand et al. 2009; Terzago et al. 2013; Schöner et al. 2019).
Other studies found decreases in spring snow water equivalent (SWE) (Figure 2-1)(Bocchiola
and Diolaiuti 2010; Marty et al. 2017a). The observed changes in snow depth and snow
duration are mainly caused by a shift from solid to liquid precipitation (Figure 2-2) and by more
frequent and more intense melt (Klein et al. 2016), both resulting from higher air temperatures
during winter and spring. These changes in snow cover duration and snow depth will have
severe consequences for winter tourism (Uhlmann et al. 2008; Steiger and Abegg 2013), water
management (Gaudard et al. 2014; Addor et al. 2014) and ecology (Wipf et al. 2009; Imperio
et al. 2013).

                                                I

Figure 2-1: The 1 Feb relative SWE deviation compared to the long-term mean (1961–
90). Annual values represent the median of the deviations of 54 in-situ Alpine stations
between 1300 and 2000 m. The smoothed red curve results from a 10-yr Gaussian low-
pass filter. (adapted from Marty et al. 2017a)

1.2 The role of snow for hydrology in Switzerland
The most important seasonal runoff signal in the Alps is the melt of snow (Beniston, 2012).
This is because the precipitation distribution is fairly even throughout the year, and because
the amount of water retained in and released from reservoirs and lakes is only a small fraction
of the total (Schaefli et al. 2007). Temperature-induced changes in streamflow (rain-to-snow
fraction, seasonal shift of snowmelt, glacier runoff contribution) are generally better understood
than the ones caused by changing precipitation (Blaschke et al., 2011). Nevertheless,
understanding long-term trends in runoff require an accurate estimate of the amount and
distribution of snow accumulation during winter (Magnusson et al., 2010; Huss et al., 2014).
The response of snowmelt to changes in temperature and precipitation is influenced by the
complex interactions between climatic conditions, topography and wind redistribution of snow
(López-Moreno et al. 2013; Lafaysse et al. 2014).
The relative share of runoff originating from snow heavily depends on the elevation of a
catchment. Since more than 50 % of the country is situated above 1000 m asl, almost 25 %
above 2000 m asl, it is not astonishing that a considerable amount of the annual precipitation
is falling as snow in the Alps. In large parts of the Swiss plateau snowfall only contributes about
5 % to the annual precipitation sum. At 1500 m asl however this ratio is already about 30%
and at 2000 m asl it increases to about 50%. The ratio is certainly varying from year to year
and it also slightly differs from region to region but is heavily depending on the elevation
distribution of an investigated catchment. During the winter season the snow pack
accumulates in many pre-alpine and most alpine regions, whereas it melts in the spring and
summer season up to altitudes of about 3,000–3,500 m asl, depending on exposure. Moreover,
the existence of a snowpack and its capacity to absorb water also controls the ratio of
precipitation which is lost by evaporation and sublimation. All these properties lead to a
seasonal runoff dynamics which is dominated by the elevation distribution of a catchment. The
role of snow as intermediate water storage for winter precipitation is of prime importance for
the understanding seasonal runoff. Low elevated catchments are usually only marginally or
only temporarily influenced by snow, whereas catchments dominated by nival regimes
                                                    2

experience the peak runoff (usually in Mai and June) with values more than double the mean
annual runoff. In glacial dominated catchments peak runoff occurs later (usually in July and
August) and the seasonal differences are even bigger due to the additional ice melt. An
example from the Brienzwiler (Aare catchment) demonstrates that the share of long-term mean
annual runoff from snow (55 %) is much larger than the share from glacier ice (13%). These
two numbers become equal (30 vs. 30 %) if only August is considered. However, the glacier
share on runoff is clearly larger than the snow share (80 vs. 20%) in the extreme case of
September 2003, following the very dry and hot summer 2003 (Stahl et al. 2017).
The seasonal differences in runoff in nival regimes are relevant in many respects: First, the
annual runoff from snowmelt is important for the management of reservoirs. Second, a
combination of snowmelt and heavy precipitation can lead to disastrous runoff and flooding,
which can cause heavy damage in alpine areas but also in downriver lowlands. Third, a low
zero degree line and therefore a larger fraction of solid precipitation (snow fall) can on the other
hand dampen the impact of heavy precipitation events. Finally, a low winter snow cover
preceding a dry spring plays an important role in the long-term forecast of summer drought.

1.3 Objectives of this chapter
The main objective of this chapter is to summarize and discuss the advances that have been
achieved in our knowledge of the role of snow on the hydrology of Switzerland in a changing
climate since the publication of the CH2014-IMPACTS report (CH2014-Impacts 2014). The
revised literature includes reports commissioned by national agencies in Switzerland and
neighbouring Alpine countries, as well as results from snow climatological studies published in
peer-reviewed journals. An additional section is dedicated to new observational technologies
that can be anticipated to play a relevant role during the next years. Finally, open research
questions are outlined and recommendations about how to translate research findings into
policy-relevant messages are provided.
This chapter is structured as follows:
   -   Review of recently published reports commissioned by national agencies.
   -   Review of recent advances in understanding and predicting future snow evolution,
       including the introduction of new technologies and techniques.
   -   Analysis of the implications of snow cover retreat on Swiss water resources.
   -   Discussion: Research gaps and open questions.
   -   Conclusions: Transform research findings into policy-relevant messages.

 Take-home messages
 Snow cover is an important water resource for the hydrology in Switzerland. Runoff
 from snowmelt plays a dominating role during one third of the year (March to June)
 in most alpine catchments.

 The high sensitivity of snow to changes in temperature and precipitation makes it a
 primary indicator of climate change and the corresponding impacts on the water
 cycle. The observed temperature increase reduces snow accumulation at low
 elevation as a greater fraction of the precipitation falls as rain, which together with
 the intensification of the snow melt causes shortening of the snow duration.

                                                 3

Figure 2-2: Changes in the ratio of snow fall days (SD) to precipitation days (PD) as a
function of altitude (m asl) of 52 Swiss observer stations for the 47 winters (DFJ)
between 1961 and 2008 (adapted from Serquet et al. 2011)

   2. Review of recent reports
2.1 Snow declines and its consequences as seen in recent reports
In recent years, several reports from national agencies, universities and research centres have
been published that investigate the impacts of climate change in Switzerland. In this section,
five major reports (see Table 1-1 in the glacier chapter) that include analyses of climate change
impacts on Swiss snow cover and their associated water resources are revised and discussed:
   A.      “Auswirkungen der Klimaänderung auf die Wasserkraftnutzung” (Schweizerische
           Gesellschaft für Hydrologie und Limnologie (SGHL) und Hydrologische
           Kommission (CHy) 2011)
   B.      “Effects of climate change on water resources and waters” (Federal Office for the
           Environment FOEN 2012)
   C.      “Assessing Climate impacts on the Quantity and quality of Water” (Beniston et al.
           2013)
   D.      “CH2014-Impacts” (CH2014-Impacts 2014)
   E.      “The snow and glacier melt components of the streamflow of the River Rhine and
           its tributaries considering the influence of climate change” (Stahl et al. 2017).
The first four reports project future impacts of climate change, whereas the fourth one presents
results of a detailed hydrological simulation of the Rhine River catchment, including an explicit
quantification of the runoff-share of snowmelt.
The snow part in report A summarized several independent studies on future runoff evolution
and on the consequence for hydropower generation from several high alpine catchments.
According to the reported results the duration of the seasonal snow cover above 1200 m asl
will shorten by 25 days per degree warming. However, the authors emphasize that the model
uncertainties are quite high at elevations between 1200 and 2000 m asl. The seasonal
                                               4

maximum SWE in the higher part of the catchments will decrease by 50 to 70 % until the end
of the century (Figure 2-3 and the variability between snow scarce and snow abundant winters
will be much smaller than today. A nowadays snow scarce winter will become normal by the
end of the century. The authors conclude that the high alpine area which is affected by rain
events will be much larger than today, especially in winter. Consequently, the runoff will
increase in winter und decrease in summer. Annual runoff from glaciated catchments may
decrease towards the end of the century due to the decreasing glacier melt. Nevertheless,
annual power production may increase due to the temporally more balanced runoff. .
Regarding snow, report B merged the result of a new study by Zappa et al. (2012) with the
summary of report A. The new study reveals that nowadays (1980-2009) about 40% of the
total runoff out of Switzerland originates from snow melt. This is about 22 km³ water per year.
Model simulations estimate that this number will drop by 40 % towards the of the century
(Figure 2-4), which would imply an annual share of only 25 % from snow melt. The projections
for the end of the century show that the snow accumulation occurs slower and the maximum
will be reached about two weeks earlier than now. An increase in winter temperatures of about
3° C by 2070–2099 will raise the snowline by around 500 metres. As precipitation amounts in
winter may increase over the course of the century, it is possible that greater amounts of snow
will fall at very high altitudes during this season. However, the total area covered by snow in
winter will become less, as will the depth of snow cover in both lower and higher regions, and
as a result there will be fewer water equivalents available to melt. This will have a direct impact
on runoff and its distribution over the course of the year. Finally, the authors conclude that the
22 km³ of water to runoff in Switzerland are considerably more than the amount contributed by
annual glacier mass loss (1 km³). As temperatures continue to rise, so will the snowline, and
regions up to about 3500 m asl will be increasingly snow-free in summer. By the end of the
century, snow cover at altitudes below 3500 m asl will last about one month less. Furthermore,
the maximum depth of the snow will decline by more than half. This will mean a reduction of
about 40% in the water reserves contained in snow cover.
Report C summarizes the results from the ACQWA-project (ACQUA 2013), which was a large
inter-disciplinary project in which the vulnerability of water resources to declining snow and ice
due to global warming was assessed. Although the project produced results for several
mountain regions in the world, here only the main results obtained for the Swiss Alps are
outlined. One of the main conclusions is that elevation-dependent impacts of warming can be
expected in the Rhone basin, with the largest impacts at high-elevation sites. Alpine snow
cover will decline due to temperature increases (more rain/less snow), particularly at mid-
latitude locations between 1000-2000 m asl. Earlier snow-melt by 5‐10 days and an increased
melt in April and May is projected for these elevations. Major changes in the total runoff of the
Rhone basin are likely to be small, but local changes or temporal changes at high-elevation
catchments can be very large. Climate change impacts in snow dominated catchments show
that water managers and users will need to adapt to changes in the quantity and timing of
water resources. This is not only relevant at local and regional scale, but also to communities
and economic sectors downstream who are reliant on a range of goods from mountain regions
and their resources (e.g. electricity, water, water storage in the form of ice and snow).
Report D on the impacts of climate change in Switzerland included a chapter “Cryospheric
aspects of climate”. In that chapter, a distributed physical snow cover model was forced by the
CH2011 climate scenarios (CH2011, 2011) to quantify how the snow cover will be affected by
climate change in upper Rhine and Aare catchment. It could be demonstrated that the
projected snow cover changes are relatively small in the near term (until 2035), in particular at
higher elevations above 2000 m asl. For instance, the spread in the projected snow cover for
this period caused by different climate model chains is larger than the difference between the
reference period and the model chain exhibiting the most moderate change. Towards the end
                                                  5

of the century, however, much larger changes with the potential to fundamentally transform
the snow dominated alpine area became apparent. These changes include a shortening of the
snow season by 5–9 weeks for the A1B scenario (Schmucki et al. 2015). This is roughly
equivalent to an elevation shift of 400–800 m. The slight increase of winter precipitation and
therefore snow fall projected in the CH2011 scenarios (with high associated uncertainty) can
no longer compensate for the effect of increasing winter temperatures even at high elevations.
In terms of SWE, the projected reduction is up to two thirds toward the end of the century. A
continuous snow cover will be restricted to a shorter time period and/or to regions at
increasingly high elevation. This implies that a multi-day snow cover is projected to become a
rare phenomenon in the Swiss Plateau assuming that the future climate is evolving according
to a non-intervention scenario like A1B or A2. The authors state that a declining snow reservoir
in the Alps will prolong periods of low river flow in summer in many parts of Europe toward the
end of this century (Farinotti et al., 2012). A significant reduction of winter snow cover will also
reduce soil moisture during dry springs and thus exacerbate the impact of hot summers.
Together with the decreasing glacier melt, this can have severe consequences for several
economic sectors including agriculture, hydropower generation, water supply and river
navigation. Furthermore, increasing river and lake temperatures may present a problem for
parts of the aquatic ecosystem (Wedekind and Küng, 2010). In the same report the chapter
“Hydrological responses to climate change” reached the following conclusions: Climate
change, and especially the higher temperatures associated with it, will cause a shift in the
runoff regime from nival to pluvial in catchments that are not strongly influenced by glacial
meltwater. Accordingly, an increase in the variability of runoff throughout the year and from
year to year, resulting in lower stability and lower predictability, is another challenge that will
need to be met in the future. A more flexible and adaptive water management will be needed
to deal with a more irregular and longer flood season (Köplin et al. 2014) and an expected
increase in the frequency of occurrence of extreme events such as droughts and floods.
Report E contains the most recent hydrological study targeting at the runoff contribution of
snow and glacier melt to the river Rhine (Stahl et al. 2017). The authors conducted a detailed
simulation of the past and current hydrology of the Rhine basin using models of different
complexity and for the period 1901 to 2006. Model complexity was chosen depending on the
characteristics of the simulated terrain along the river basin, and rain, snow- and ice-melt
contributions to runoff were simulated explicitly. One important result considering past climate
conditions was the large relevance of snowmelt along the Rhine River. The mean annual
fraction of snowmelt in Basel was modelled as 39 %, compared to only 2% originating from
glacier melt. These numbers agree well with estimates in previous modelling studies. Thanks
to the daily resolution of the modelling this study could, for the first time, specifically quantify
the snow and ice melt component during extreme low flow years such as 1921, 1947, 2003.
While the maximum daily fraction of glacier melt during summer 2003 (i.e. end of August) was
25 %, the mean snowmelt fraction during the same time was still 20 %, demonstrating the
significant role of the snow cover for runoff even after a hot and dry summer. Therefore, the
authors conclude that this highly relevant contribution of snowmelt to streamflow suggests that
changes in snow cover and timing of snowmelt as a contribution to runoff will play a crucial
role in the future.

2.2 Open questions in the reviewed reports
By far the largest question mark in all reports is the development of future precipitation. In
contrast to temperature projections, projections for precipitation still show a large spread
between different climate model chains. Both increasing and decreasing precipitation is found
for the Alps, and also the spread between different regions is large. Modelled changes are

                                                 6

small for the near future (2035), but can range between +30 % and -20 % for extreme scenarios
towards the end of the century. Since at least part of this precipitation will fall as snow, the
future mountain snow cover will be affected accordingly. The uncertainty of the projections
triggers the question whether the anticipated increase in winter precipitation will be able to
compensate for the expected reduction in snowmelt runoff in spring. So far, another important
limitation is the fact the most projections are based on the so-called delta change method
(Bosshard et al. 2011), which assumes that future inter-annual variability is unchanged.

 Take-home messages
 During the period 2010-2014, a series of reports targeting at climate change impacts
 on water resources were commissioned by Swiss national agencies. The
 corresponding studies have included numerical simulations of future snow changes
 and their contribution to runoff.

 All studies project significant elevation dependent reductions (between 50% and 97%)
 in the snow volume by 2100 assuming the A2 emission scenario. However, the large
 spread in the projected temporal and spatial precipitation evolution by the different
 climate models produces relatively large uncertainties.

 A recent report commissioned by the International Commission for the Hydrology of
 the Rhine River incorporated the effect of snow cover in several hydrological models
 for the Rhine River basin. Runoff from snowmelt was determined to provide a
 significant contribution to the catchment runoff, even during hot and dry summer
 months as last experienced in 2003. In Basel, for example, the mean annual fraction
 of snowmelt was modelled as 39%, compared to only 2% originating from glacier melt.

Figure 2-3: Snow covered area as function of time of the year (left) and distribution of
seasonal maximum SWE as function of altitude (right) for today (solid line) and for the
end of the century (2071-2100) and the IPCC A1B (grey dashed) and the A2 (black
dashed) scenario in the Damma-Reuss catchment (adapted from Magnusson et al.
2010).

                                               7

Figure 2-4: Projected decrease of meltwater from snow shown as a percentage of total
runoff for all large catchments in Switzerland towards mid and end of the century for
the A1B scenario (adapted from Zappa et al. 2012)

    3. Progress in understanding and predicting future snow evolution
3.1 State of literature
A longer sequence of low snow winters in the Alps, as it was observed between the late 1980’s
and mid 1990’s, led to the question if this was caused by climate change or natural variability.
The only study published during this period had to limit its analysis to a few stations and to the
50 years between 1945 and 1994 (Beniston et al. 1997), and it was realized that the available
snow data was insufficient to prove the uniqueness of such a sequence. A national research
program on climate change and natural hazards (NRP31) in the mid 1990’s allowed for
digitization of archived snow data, combination of overlapping series, and quality-checking of
already available data series. This work notably enabled two studies, investigating variability
and trends in archived snow-cover data (Laternser and Schneebeli, 2003; Scherrer et al.,
2004). Both studies found a decrease of the snow depth since the mid 1980’s for low-lying
stations. This decline could be linked to anomalous warm winter temperatures and more
intensive spring snowmelt (Marty and Meister 2012). The step-like increase of temperatures
could be associated with (a) the longest observed period with a strong, persistent, positive
North Atlantic oscillation (NAO) index and (b) series of extended blocking high-pressure events
(Scherrer and Appenzeller 2006; Marty et al. 2008). In terms of implications, these studies
focused mainly on the impact of snow depth or snow duration changes on winter tourism and
less on water resources.

                                                8

Most of the snow-hydrological analyses related to climate change were carried out in the frame
of three projects: CCHydro (funded by the Swiss Federal Office for the Environment), FUGE
(funded by the Swiss National Foundation in the frame of the National Research Programme
61) and ACQWA (funded within the 7th framework program of the European Union). The
projects were all dealing with water and climate change in general, but also analysed the role
of snow as an important water resource in Switzerland. The projects were an important basis
for the four reports mentioned in the former chapter. An additional project, founded by the
European Union Climate Change Adaptation by Spatial Planning in the Alpine Space (CLISP)
program, had a partial focus on snow resources as well. In fact, future projections for Alpine
water resources based on different scenarios were performed in high spatial resolution for the
Rhine and Inn catchment (Bavay et al. 2013).
As a result of this intense scientific activity, the future of the Swiss snow cover has been
increasingly included in the scientific discussions about climate change. Facilitated by the
availability of high-quality observations for verification, moreover, case studies in the Alps are
frequently used as a reference for other regions. All these studies demonstrate a clear impact
of rising temperatures on snow cover depth and duration Since the evolution of the snow cover
at elevations below 1400 m is dominated by temperature (Moran et al. 2013) the projected
response has much smaller uncertainties at low elevation than at high elevation. For the latter,
in fact, precipitation variability has a non-negligible impact. Even if the general consequences
of climate change on the Swiss snow cover are consistent within the mentioned studies,
significant uncertainties and differences are still found. This is mainly because of divergent
climate scenarios or different strategies for the preparation of future meteorological data as
input for the snow models.

3.2 Recent advances
A recent study by Klein et al. (2016) investigated past trends of snow duration and maximum
snow depth over the period 1970–2015 in the Swiss Alps, based on eleven stations between
1139 and 2540 m asl. The results show a clear reduction of the snowpack, irrespective of
elevation or region, and a clear shortening of the continuous snow cover duration (-8.9 days
per decade on average). The reduction in snow cover duration is mostly the result of earlier
snowmelt (on average by 5.8 days per decade), rather than of later snow onset. The authors
also found a general decline in the annual maximum snow depth. Moreover, the time of
maximum snow depth also occurred earlier, and was highly connected to the snowmelt date
and much less to the time of snow onset, illustrating the impact of the spring temperatures and
of the earlier start of the melting period on the snow season.
Similar to the afore-mentioned study, past snow trends were so far always analyzed based on
snow fall or snow depth. A new study by Marty et al. (2017a) analyzed for the first time long-
term SWE data series and demonstrates that the Alpine snow water resources have been
decreasing during the last six decades at least (Figure 2-5). This is true for all elevations,
including the highest ones. A spatial analysis revealed that strength or significance of the SWE
reduction is independent of the geographical position within the Alps. Low-elevation stations
less frequently demonstrate significant trends than higher ones, due to the higher inter-annual
variability. The observed decreases are clearly larger for April than for February. Trends in
April also show a clear elevation dependency. For the last 45 years, for example, relative April
SWE-loss ranges from 80% at lowest elevations (

Regarding the future evolution of snow water resources in Switzerland, two new studies
demonstrate that runoff is mostly affected by the decrease in continuous snow cover and in
the mean snow depth, especially if dry weather conditions prevail. The projected relative
reductions of SWE towards the end of the century for the A1B scenario are in the order of 35%,
85% and 95% at 2500 m, 1500 m and 500 m elevation, respectively (Schmucki et al. 2015),
which drastically reduces the amount of snow mass available for runoff. The second study
(Marty et al. 2017b) used a wide range of climate and emission scenarios to show that the
duration and mass of the snow cover in typical Alpine catchments such as the Aare (Figure 2-
6) and Rhine will shrink until the end of the century independently of the emission scenario
and the climate model used. The authors also demonstrate that the snow loss could be
reduced by about 40% if global warming was to be limited to 2 degrees. The results of these
two studies are corroborated by the widespread projected trend of decreasing snowfall in
ensemble of the newest generation of RCMs (Frei et al. 2018) and clearly decreasing trends
of snow depth in the French Alps (Verfaillie et al. 2018)
Both of these studies applied a delta-change approach to produce meteorological time series
of the future, despite the fact that this method is not capable of describing possible changes in
temporal variability. A recent study by Keller at al. (2017) demonstrated that the alternative
method of using a weather generator does not generally give different results as long as the
summer season with the projected large change in the dry-wet structure is not included.
The question how strong maximum winter SWE impacts summer low flow was investigated by
Jenicek et al. (2016). The results showed that a decrease of maximum SWE by 10 % results
in a decrease of minimum discharge in July by 6–9 % in catchments higher than 2000 m asl.
This effect was smaller in middle- and lower-elevation catchments with a decrease of minimum
discharge by 2–5 % per 10 % decrease of maximum SWE. The authors also demonstrated a
higher sensitivity of summer low flow to snow accumulation in Alpine catchments compared to
lower-elevation pre-Alpine catchments.
A recent micro-meteorological study in an Alpine catchment demonstrated that turbulent
sensible heat fluxes over typical melting patchy spring snow cover snow are strongly
overestimated in all energy balance models (Mott et al. 2017). This influences the heat
exchange between the snow cover and the overlying atmosphere and has strong implications
for hydrological modelling covering an entire ablation season.
Within the CH2018 initiative, new climate scenarios for Switzerland have been produced
(CH2018, 2018). These scenarios are based on the latest set of simulations for the European
climate from the Coordinated Regional Climate Downscaling Experiment (CORDEX). The new
set of scenarios provide the standard input variables for future climate change studies in
Switzerland, including those addressing future snow changes and related impacts.

3.3 New technologies and remote sensing products
Complementary to advances in climate and snow modelling, some important advances have
recently been achieved in the domain of monitoring techniques. These technologies can be
expected to be important in the validation and further development of modelling approaches,
ultimately helping to reduce the uncertainties of future projections.
New techniques are in development in the domain of in-situ monitoring of snow depth and
SWE. A new multipoint laser snow depth sensor helps to close the gap between standard ultra-
sonic sensors and the single point laser sensor (Kim et al. 2014) Low-cost GPS sensors can
be used to detect snow depth and SWE simultaneously (Koch et al. 2019). Electronic handheld
devices like a self-recording snow depth probe (Sturm 1999) or a high-resolution penetrometer
                                               10

for snow, the SnowMicroPen (Proksch et al. 2015) have been developed to efficiently
investigate the small-scale heterogeneity of snow depth or snow density. A combination of a
cosmic ray sensor and weather radar composites have been successfully applied to spatially
integrate SWE over a Swiss glacier (Gugerli et al. 2019). High-resolution monitoring of snow
surfaces by laser scanning has been successfully applied to map the snow distribution. The
technique allows for detailed DEMs to be acquired with frequent time intervals and the
necessary instruments can be either terrestrial (Bühler et al. 2009) or aerial (Gruenewald and
Lehning, 2011). The relatively high costs of the instruments, the restricted spatial range and
temporal availability have, however, limited the widespread extension of the technique. The
use of close-range photogrammetry based on Unmanned Aerial Vehicles (UAVs) has lowered
the cost of obtaining repeated DEMs but is still restricted in regard to the area and temporal
availability (Bühler et al. 2016).
In addition to classical remote sensing products, some satellite records are becoming long
enough to be used for climatological studies (Hüsler et al. 2014). Moreover, a recent
generation of satellite products (ESA Sentinel) is anticipated to significantly improve the
monitoring of snow area and snow mass. This will produce more accurate validation datasets
for snow models operating at the larger scale, thus fostering their development (Lievens et al.
2019).

 Take-home messages
 Past trends in Alpine snow water resources have recently been studied on the basis
 of long-term snow water equivalent measurements. The results show a clear decline,
 especially in spring. The loss ranges from 80 % below 1000 m asl to about 10 % at
 2500 m asl for April 1 since the 1960’s.

 The research focus has recently targeted future projections of snow water resources.
 The projected relative reductions of the mean SWE towards the end of the century in
 a typical large Alpine catchment are in the order of 70 % for the A2 scenario.

 New monitoring techniques, such as UAVs or terrestrial and airborne laser scanning,
 but also new satellite products with increased spatial and temporal resolution will
 play an important role in the years to come. The acquired data can be anticipated to
 foster model improvements and better process understanding.

                                              11

Figure 2-5: April 1 relative (left) and absolute (right) SWE trends (1968–2012) of 54
long-term stations in the Alps as a function of altitude. Large triangles indicate
significant trends (p < 0.05) and small triangles indicate weakly significant trends (p <
0.2). Circles represent stations with no significant trend (p ≥ 0.2) (adapted from Marty
et al. 2017a).

Figure 2-6: Mean (solid), maximum and minimum (dotted) snow depth for the reference
period (black) and the A2 scenario (blue) towards the end of the century in the Aare
catchment for six elevation zones. The elevation zones are 100m wide, i.e., the 1500m

                                           12

zone, contains all pixels between 1450 and 1550 m asl. Note that the scale of the y axis
changes with elevation (adapted from Marty et al. 2017b).

   4. Implications of projected snow changes on Swiss water resources
4.1 Water availability
Challenges already persist in the governance and management of water availability. Climate
change impacts are likely to exacerbate these issues, by introducing an extra layer of
uncertainty and shifting the hydrological baselines upon which rules and policies are based for
different scales and sectors (Clarvis et al. 2014). Most ski resorts invested millions of Swiss
francs in the last two decades in a network of machines to produce technical snow along their
kilometre long ski runs in order to gain a competitive advantage and to be prepared for snow
poor winters. Such a network requires vast of water either from artificial reservoirs on the
mountain slopes or from rivers or lakes in the valleys. This water is needed in times when
natural flow is already on its lowest level and when Alpine winter tourist areas have the highest
demand. This is due to both artificial snow production and the doubled or tripled number of
people during the Christmas/New Year holidays. After a dry fall such situation have locally
already initiated discussions about the correct governance of water use (Schneider et al.,
2015). However, this particular problem may become slightly smaller in the future due to an
expected increase in winter runoff from Alpine catchments in a warmer climate (Bavay et al.
2013). On the other hand less snow in spring increases the probability for low flows in summer.
Such changes in the seasonal availability of water resources can have local and temporal
impacts on water management, such as irrigation of residential zones, drinking water or water
uses related to tourism (e.g., artificial snow production, irrigation of golf courses) or to
agriculture, especially in southwestern Switzerland (Reynard et al. 2014). Although there is no
water stress in this region at the moment, it can be assumed that through the general decrease
in SWE, less water will be available during summer, where water demand for irrigation is
increasing (Jenizek et al. 2018, Etter et al. 2017). However, the size and temporal shift of the
seasonal change is highly dependent on the hydrology of the region, i.e. rain-dominated
catchments react differently than melt-dominated catchments, especially for extreme flow
regimes (Brunner et al. 2019).

4.2 Hydropower production
The increasing production of artificial snow needs not only water, but also power. The
necessary water for the snow production often needs to be pumped up from rivers or
hydropower reservoirs. Local reservoirs built to store water for snow production or usually too
small and have to be filled several times during the early winter. This happens at times when
natural river flow is at its lowest level, which together with the demand for hydropower
production can cause problems with environmental regulations. The activity can also be cost-
intensive, since electricity prices are higher during winter due to generally increased demand
in that season. The presence of snow in the Swiss Alps has therefore a direct bearing on the
energy sector. This is important in countries like Austria or Switzerland that depend for more
than 50 % on hydropower for electricity generation (Romerio 2002). The projected decrease
in snow depth and snow duration will drastically reduce the amount of snow mass available for
runoff. Accordingly, peak runoff from snow is projected to occur earlier in the season or might
even be suppressed in some years. Therefore, it can be assumed that the electricity production

                                               13

through hydropower will be increased during the winter season, but reduced in summer due to
the large reduction in snow cover. This implies that runoff is less variable through the year,
which may positively affect hydropower. On the other end, the abovementioned reduction in
summer runoff implies that some mountain rivers may no longer be able to maintain sustained
discharge during prolonged dry spells (Addor et al. 2014; Köplin et al. 2014).

4.3 Natural hazards and other impacts
The above-mentioned intensification of snowmelt, the increase in rain events, and the general
decrease in snow depth and snow duration will probably affect the frequency, number and
seasonality of landslide activations and debris flows (Stoffel et al. 2014). Moreover, rapidly
melting snow can, in the presence of heavy rainfall, provide significant amounts of water that
converge towards mountain torrents and their tributaries, leading to floods in some instances
(Stoffel and Corona 2018). The future frequency and intensity of hazardous runoff from rain-
on-snow events is uncertain (Wuerzer and Jonas, 2018), but according to Beniston and Stoffel
(2016) rain-on-snow may increase in future despite a generally reduced snow cover as long
as the warming does not go beyond +4 degrees. The same processes may also enhance the
risk for wet snow avalanches (Pielmeier et al. 2013). Moreover, more frequent and especially
more intensive winter rainfall will probably increase the risk for mixed avalanches. These are
avalanches which may start as a power snow avalanche on the top of mountain and end as a
mixture of a wet snow avalanche and a debris- or mud flow in the valley ground. This type of
avalanches often constitutes a problem since the amount of water reduces the friction, which
can dramatically increase the runout distance of the avalanche. On the other end, the projected
decrease in the number of days with snowfall will reduce cost for snow removal of
transportation infrastructure (Figure 2-7), for road accidents, for airport and road closures.
Other well-known impacts comprise winter tourism, which is the main economic sector in many
Alpine valleys. This business is among others impacted by climate change, i.e., heavily
dependent on snow (Gonseth 2013). Ski resorts are highly affected by the projected decrease
in snow depth and snow duration, since the absence of enough natural snow needs to be
compensated through the production of artificial snow (Figure 2-8). Besides being highly
temperature dependent, this also implies higher water consumption, and is very cost-intensive
(Abegg 2013). It can be assumed that the small ski resorts at mid elevations, which are already
endangered today, will disappear. The remaining resorts are projected to experience a non-
continuous snow cover at least every second winter. The strong decrease in snow duration,
i.e. the earlier snow disappearance may also changes the visual appearance of the mountains
during summer and thus heavily influencing the “Alpine landscape” perception.. For example,
the typical summer tourist picture, snow covered peaks in front of green meadow, may become
hard to get after a few hot weeks in a future summer.,
A shorter snow duration, especially in spring and summer also heavily influences the glacier
melting and the permafrost warming (Huss et al. 2017), because the reflecting surface of the
snow cover cannot anymore protect against the strong summer sun (see glacier- or permafrost
chapter, respectively). Even flora and fauna do not necessary profit from a longer snow free
period (Ragora et al. 2018, Penczykowski et al. 2016). Some species need the snow to protect
themselves from freezing temperatures or to hide from predators. Some species may be able
to migrate to higher ground others may become extinct due to their too slow reaction or due to
a missing appropriate habitat.

                                              14

Take-home messages
 Future declining snow water resources in combination with dry summer periods can
 locally and temporarily limit the water availability towards the end of the century.
 However, improved governance and management of the available water can largely
 mitigate the impact of such situations.

 Hydropower production will be mainly affected by changes in seasonality. Due to a
 shift in water availability, production revenue may increase during the rainier winter
 months and decrease during the drier summer months.

 Snow cover decline will also have other important consequences, such as negative
 effects on winter tourism or a possible increase in wet snow avalanches. But there
 are also positive effects like reduced cost for winter road maintenance, less snow
 induced traffic interruptions or road accidents.

Figure 2-7: Annual mean de-icing salt consumption in Switzerland for the reference
period 1997-2012 and its 50 % range (black), and the three future periods 2020-49,
2045-74 and 2070-99 with different emission scenarios (colored). The impact of the
A1B (violet), A2 (grey) and the RCP3PD (yellow) scenario are given from left to right
(adapted from Zubler et al. 2015).

                                           15